|
|
|
#46
|
||||
|
||||
Hey good progress Shern. You’ve gotten some great info. Basically it’s one of two things. Either the knob has too much residual resistance when it is supposed to read zero or the thermistor is reading too high. Both components can drift with age.
The thermistor is definitely NTC. We know this because when you short the thermistor input on the knob assembly it turns on your AC. That means hot temperature is low resistance. Open shuts things down, which is high resistance means too cold. You are probably out of cal but you may not be using the same thermistor as the one I show for the 300D. That’s a totally different application. It’s in room temp air. Yours is inside the air box. Granted your result of reading 18K at room temp is higher than 10K and reducing the expected thermistor resistance by shorting it with a wire (reducing the resistance) says taking it down from 18K to 10k may be the right move. There are a couple of things you can do. The idiot thing is wire a switch across those two terminals you show in your image. This will mean when the switch is open you’ll get your 46-62F temps while driving on max. Then when you need really cold you throw the switch and get your 37F. The more sophisticated way is trim the value of the thermistor by putting a big resistor in parallel with it. Like if you wanted to cut the 18K in half you put another 18K in parallel with it and it’ll be 9K. It looks like that is what somebody did already? That looks like an 82K in parallel with whatever you have in the air box. Sooo…that’s 14.76k at room temp. Edit: I think I got that wrong, it should be 23.4k at room temp in parallel with 82k which gives 18K measured. Looks like that’s a trim resistor. You may want to reduce its value a bit and see what happens. One easy way to empirically map out the desired set point is replace that resistor with a 100K potentiometer and dial in your desired temp range. Start at 100K and work your way down. It’ll likely be in its optimum spot somewhere between 82K and zero(the point you got 37F air). Make sense? I think I have that right. You want to reduce the resistance of the thermistor to make the Temperature Control think it’s hotter than it is so it turns on your compressor. I can draw a picture if you need a diagram. Then take off the potentiometer without changing the magic setting. Measure its resistance with your meter. And replace that nominal value with a fixed resistor. Basically you’re reducing the value of that 82K trim a touch. Hats off to chasing this down. I think you’re almost at the point of having a knockout AC system. If I think of anything else I’ll post an edit.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles Last edited by ykobayashi; 05-05-2022 at 11:44 PM. |
#47
|
||||
|
||||
Fascinating
I'm reading all this and wondering if I'll ever get my '82 240D's AC up to snuff .
__________________
-Nate 1982 240D 408,XXX miles Ignorance is the mother of suspicion and fear is the father I did then what I knew how to do ~ now that I know better I do better |
#48
|
||||
|
||||
Thank you so much for this. Chatted with H3ffe this morning and confirmed the resistor is actually oem. His reading however clocked in at 14k-ish… I am definitely out of spec.
Been hunting around a little the past few days for a new sensor (laughably NLA) but managed to find a single unit -apparently it’s been on the shelf for three years. That of course beats 42 years in service. I bought it. Should arrive end of next week. Once it’s installed I’ll see where I’m at. The potential for additional drift within the AC wheel is on my mind, though I can only diffuse one bomb at a time… Will update the thread once it’s live. It’s odd to me that this hasn’t come up before. Granted, I’m not sure how many manual hvac systems are in action -typically 240ds and there aren’t many of us left- but thought I’d have encountered at least one single mention online. Nada. We’re pioneering here I suppose. Should that part not arrive for whatever reason, the resistor fix might become standard practice.
__________________
1981 240D 4sp manual. Ivory White. |
#49
|
||||
|
||||
Resistor Value
Isn't the bands of color supposed to tell you what exact resistance it is ? .
I took some radio building courses in the mid 1960's and dimly recall this..... Maybe you could decipher the original value and replace it ? .
__________________
-Nate 1982 240D 408,XXX miles Ignorance is the mother of suspicion and fear is the father I did then what I knew how to do ~ now that I know better I do better |
#50
|
||||
|
||||
Possibly, but that would only be the resistor.
I’d still have the thermistor to contend with.
__________________
1981 240D 4sp manual. Ivory White. |
#51
|
||||
|
||||
The bands say 82K.
No Shern, you misunderstood. We change that little 82K trim to bring your system into spec. Leave your thermistor alone. It works. Don’t buy anything. You just need to reduce the value of that 82K trim resistor. It’s a 5 cent fix. The trim is like the little knob on my wife’s bathroom scale. I can turn the knob and it moves the dial below zero. She can still weigh herself but she will be happy because she’ll be ten pounds lighter across the board no matter what she weighs on that day. And then we’ll get to eat desserts again. I think you just gave me enough info from H3ffe’s 14K value. I can solve for the trim resistor value you need to buy. Check back later. It’s too early for my brain.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles |
#52
|
||||
|
||||
Ah! Okay.
In that case, standing by! I thought the thermistor itself (the part you linked on mouser) was also subject to age drift. I should also note that we’re not sure whether h3ffe’s sensor is in spec. He’s currently rebuilding his entire system. The part isn’t bad -$16. And I can also return it. Might not be bad to measure it and establish a baseline? Edit: brain also coming back online. If it’s just the resistor, couldn’t I simply replace it one for one?
__________________
1981 240D 4sp manual. Ivory White. Last edited by Shern; 05-07-2022 at 01:10 PM. |
#53
|
||||
|
||||
Yes, just desolder and change the little trim resistor. Leave the thermistor in the airbox alone.
33k, 1/8w, 5% resistor. Pick up a few other values like 18k, 22K, 24k in case you want it colder. ETA - lots of good electronics surpluses up in your area. I recall ALL Electronics but there will be others who stock these resistors. All thermistors drift a fraction of a degree a year. After 40 years it adds up. You can buy the $16 part…but why. It’ll undoubtedly be trimmed too. I think MB did this at the factory to make their CCUs “perfect” kind of like shimming the door latches. This is really easy and cheap. The hard part is done. Oh yeah and I forgot to say h3ffe may be off spec too. But he’s definitely going in the right direction. A 14K total resistance will give you a lower shutoff temperature. That’s why i say get a few other values in case it needs to be further reduced.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles Last edited by ykobayashi; 05-07-2022 at 01:51 PM. |
#54
|
||||
|
||||
Quote:
Other people are reading this thread. There will be a demand for that part once people realize they aren’t hitting their lowest temperatures because of a $16 part. Reminds me of this clip. We don’t even have to bury it 1000 years like Belloq. https://youtu.be/1Hx05Ia_byQ
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles |
#55
|
||||
|
||||
Quote:
Lol it’s time for a re-watch. Excellent diagram, thank you. One of the first repairs I ever did on this car was the old clock fix with a new capacitor. To this day, reconnecting the battery and hearing that heavy “tick… tick… tick” was one of the most satisfying repairs I’ve ever made. I’ll pick up some resistors and have at it.
__________________
1981 240D 4sp manual. Ivory White. |
#56
|
||||
|
||||
It looks like All Electronics in Van Nuys has the goods. Or some version of them. Oddly this was the only local shop I could find that carried what I needed. Unless these businesses are google proof -which wouldn’t surprise me- there aren’t too many left.
Couple layman questions for you… 1.) can I use multiple resistors in series to get the values I need? Using your equations it appears that I can, but my understanding of electrical engineering is linear. Much more variety in the 1/6, 1/4w style. I.e. I can’t find 33k, but I can find 24k and 10k. 2.) 1/8 watt seems critical, but is the tolerance value of 5%? I’ve found a few kits on eBay and Amazon with 1%. In general I’d imagine the lower the tolerance value the better, but maybe in this case 5% creates a smoother gradient? (Again, linear understanding…) 3.) if im using resistors in series, is the watt value cumulative? (I assume not, but hey). Tbh i’m not even sure if I’ll be able to close the wire harness with more than one resistor installed.
__________________
1981 240D 4sp manual. Ivory White. |
#57
|
||||
|
||||
Hi Shern,
Almost any kind of resistor will work. The reason to use 1/8W 5% was that is what is there right now so I figured that would fit best. 1). Yes. Multiple in series will work. Yes, 24K + 10K = 34K. Also 68K in parallel with 68K gives 34K. 2) 1% will work fine. No smoother with either one. I just picked 5% because that’s what’s there. 3) yes. Each resistor can dissipate an 1/8W. If you distribute the power over two you can distribute 1/4. But I don’t think it’s an issue here. We aren’t burning much power. Closing the wiring harness is the trick. I don’t think you have to get that close to 33K. You’re already way off and the circuit kind of works. Maybe the magic number is closer to 24K. Maybe it’s 47K. 33K is just to trim you to the same value that H3ffe had. 33K is a standard value. It should be easy to find. Huh…not at ALL Elec. I see. Ok. Maybe this is just what they list. My recollection is the place is a big jumble sale and they may have kit boxes you can buy one resistor of a set value from. I wouldn’t go by what they list. Just go up and ask with the understanding that almost anything close 20K - 68K will work in 1/8-1/4w. Ask. And as I said before, if you want to get really tricky buy a small 100K potentiometer. With this you can dial in resistances from 0-100K. You can dial around till you find a value you like. https://www.allelectronics.com/category/530250/resistors/1/8-watt-resistors/1.html You can try a 20K, 27K and 47K. Close enough I think. Or you can try the 33K in 1/4w. It’s bigger but maybe you can bend it over like an old Japanese radio mount. (Like VDO does it in our cars). Example. To save space they are bent into a U and placed upright. Manipulating Electricity by Storm Crypt, on Flickr Vintage Radio Electronics by BlazerMan, on Flickr ETA- yeah like this’ll work. You get one of these and it’s alike having all resistors from 0-100K. Put a couple of long wires on it and tie it in in place of the 82K. Use the middle terminal and one of the edge terminals for your two connections. I’d set it at 82K for starters (set it with a meter) then solder it in. Then you adjust and watch the switch points go up and down on your thermometer. When you’re happy stop. https://www.allelectronics.com/item/nltp-100k/100k-linear-taper-pot-6mm-shaft/1.html Take out the pot. Measure it. Order the exact 1/8W leaded resistor on eBay from China. Put it in when it comes a month later. There are many ways to do this.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles |
#58
|
||||
|
||||
Okay, wonderful... thank you for this.
I really do like the idea of installing a potentiometer to dial in the right value. This would certainly make the trial and error process much less awkward. To get to a safe evaporator cutoff point (let's say 38F -if I recall, vent temps are 5F higher than the evaporator?), I'll need to be traveling at highway speed -the coldest it'll get. At which point, I'll need to gradually dial up the resistance until the next lowest setting under "max" cycles off the compressor. Keep an eye on temps, dial down the temp wheel some more, see what kind of a temperature gradient I'm getting. I guess at that point, I pull over and measure resistance and learn what resistor I need. Come to think of it, this would be a lot easier to trouble shoot in the winter while parked in the driveway. Re: the pot, does this look like the guy for the job? https://www.allelectronics.com/item/nltp-100k/100k-linear-taper-pot-6mm-shaft/1.html The only thing I'm not clear on, is which of the three prongs I should use?
__________________
1981 240D 4sp manual. Ivory White. |
#59
|
||||
|
||||
Yeah any of the methods will likely work. Driving around I bottom out at 37F at the vent on my 300D. That’s compressor on all the time with no monovalve mixing in. On your manual setup it’ll probably be best if you hit 37F and you see it cut off and it drifts up to some temperature in the forties then turns on again. This is all kind of an experiment. My gut feel is if my ETR is cutting at 37F (vent) that means the evaporator is close to freezing and MB designed it to stop there. Maybe you can go to 35? 34? I dunno. At some point your evaporator will freeze up I guess. I don’t think there is a terrible penalty for freezing up your evaporator. Somebody chime in if I’m wrong here. It’ll just stop cooling well till the ice melts away.
I’ve done it a few times in my truck which has a very old .(1950s) thermostat that lets me set too low a temperature. It hasn’t seemed to hurt it. I just remember not to turn the setting to the lowest level with the fan on low. It kind of stops cooling as it freezes. This is all kind of an experiment. Yeah that’s the pot you want. Pick one outer terminal and the inner. Depending on which outer you pick the resistance will increase or decrease as you turn clockwise. It’ll work either way but one way will change in the reverse direction. The symbol for a potentiometer makes it clear why. At the extremes the knob makes the connection between the two terminals a wire or a 100K resistor. When in doubt check it with a meter. Pick one side. One connection has to be the arrow (“wiper”) or it’ll just be a fixed 100K resistance between the two ends. One end, one center and you’re good. You can use a meter of wire so you can comfortably adjust the pot as you drive around. Set it at 82k before you solder it in. Then solder it in. Run the AC. Dial it down as you measure temperature. In theory it should be around 33K. When you’re happy with the temps, remove the potentiometer without spinning the knob. Measure it. That’s your fixed trim. Buy that value and solder in in. It’s gonna be funny if none of this works. There’s theory and reality. I think we are on the right track. You’ve technically tried three resistors. Infinity (unplugged). Zero (shorted with a jumper) and 82k. It has to be somewhere between 0 and 82k.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles |
#60
|
||||
|
||||
Electronics Stores
Try Hawthorne Electronics on Hawthorne Blvd., off the i110 freeway go West, it's on the North side .
Not a lot of old timey electronics stores like decades go . A really nice one in Fullerton, I forget the name though .
__________________
-Nate 1982 240D 408,XXX miles Ignorance is the mother of suspicion and fear is the father I did then what I knew how to do ~ now that I know better I do better |
Bookmarks |
Thread Tools | |
Display Modes | |
|
|